Closing a conceptual gap: the case for a description of ...

A DESCRIPTION OF ENGLISH AS A LINGUA FRANCA 133

© Blackwell Publishers Ltd. 2001© Blackwell Publishers Ltd. 2001, 108 Cowley Road, Oxford OX4 1JF, UK and350 Main Sheet, Malden, MA 02148, USA

INTERNATIONAL JOURNAL OF APPLIED LINGUISTICS, Vol. 11, No. 2, 2001 133

Closing a conceptual gap:the case for a description of

English as a lingua franca

BARBARA SEIDLHOFER

University of Vienna

Despite momentous developments in the sociopolitics of the teachingof English worldwide, targets have generally remained tied to native-speaker norms. This paper argues that although this orientation isoften recognized as inappropriate and counter-productive, it persistsbecause discussions about ‘global English’ on the meta-level have notbeen accompanied by a necessary reorientation in linguistic research:very little empirical work has so far been done on the most extensivecontemporary use of English worldwide, namely English as a linguafranca, largely among ‘non-native’ speakers. The paper seeks to demon-strate that this lack of a descriptive reality precludes us from con-ceiving of speakers of lingua franca English as language users in theirown right and thus makes it difficult to counteract the reproductionof native English dominance. To remedy this situation, a researchagenda is proposed which accords lingua franca English a centralplace in description alongside English as a native language, and a newcorpus project is described which constitutes a first step in this pro-cess. The paper concludes with a consideration of the potentially verysignificant impact that the availability of an alternative model for theteaching of English as a lingua franca would have for pedagogy andteacher education.

Introduction

Fundamental issues to do with the global spread and use of English have, atlong last, become an important focus of research in applied linguistics. Thedebate has been conducted particularly vigorously over the last decade, vari-ously highlighting crucial cultural, ecological, socio-political and psychologicalissues. The realization that the majority of uses of English occur in contexts

134 BARBARA SEIDLHOFER

© Blackwell Publishers Ltd. 2001

where it serves as a lingua franca, far removed from its native speakers’linguacultural norms and identities, has been an important leitmotif in thisdiscussion.1 And yet, the daily practices of most of the millions of teachers ofEnglish worldwide seem to remain untouched by this development: very fewteachers ‘on the ground’ take part in this meta-level discussion, and most class-room language teaching per se has changed remarkably little considering howthe discourse about it has. This is not surprising if we consider that what theseteachers of English generally regard, for better or for worse, as their mainknowledge base and point of reference, the target language as codified in gram-mars, dictionaries and textbooks, has not moved with the tide of applied lin-guistics research. This state of affairs has resulted in a conceptual gap in thediscourse of ELT which, although it is giving rise to some misgivings and un-ease, has not been addressed directly and proactively. This paper attempts totake stock of this unsatisfactory situation and proposes a project that mayconstitute a radical but promising way forward.

A conceptual gap

Teaching English, which well into the heyday of communicative language teach-ing seemed a fairly straightforward activity, has become a much more complic-ated affair. Whereas language teachers used to be mainly educated about andpreoccupied with various approaches to the description and instruction of thetarget language as such, we now find a much wider variety of concerns with thekind of socio-political and other issues referred to above demanding at least asmuch attention as the language proper. This has led to a broader conception ofthe profession, and to a discourse of ELT in which notions of ‘correctness’,‘norms’, ‘mistakes’ and ‘authority’ seem to have given way to an ethos char-acterized by ‘transformative pedagogy’, ‘learner-centredness’, ‘awareness’and ‘(self-)reflection’ (cf. e.g. Pennycook 1999). In the discourse of languageplanning and education policy, monoculturalism, monolingualism, monomodelsand monocentrism have been replaced by multiculturalism, multilingualism,polymodels and pluricentrism (cf. e.g. Bamgbo7e, Banjo & Thomas 1995;Bhatia 1997; Kachru 1992b; McArthur 1998; Smith & Forman 1997). Euro-pean sociolinguists have expressed their concern about “English only? inEurope” (Ammon, Mattheier & Nelde 1994) and second language acquisitionresearch has taken its first significant steps “beyond the native speaker” (Cook1999).

The most important consequence of these developments for so-called non-native teachers of English, the majority of teachers of English worldwide, hasprobably been that the notion of native speakers’ “ownership of English” hasbeen radically called into question (Widdowson 1994) and that a discussionhas gathered momentum which highlights the potential special expertise ‘non-native’ teachers have on the grounds that they know the target language as aforeign language, share with their students the experience of what it is like to



try and make it their own, often through the same first language/culture ‘filter’,and can represent relevant role models for learners (cf. e.g. Braine 1999; Brutt-Griffler & Samimy 1999; Kramsch 1998; Medgyes 1994; Rampton 1990;Seidlhofer 1999).2

The whole orientation of TEFL (teaching English as a foreign language),then, seems to have fundamentally shifted: from correctness to appropriate-ness, from parochial domesticity and exclusive native-speaker norms to globalinclusiveness and egalitarian licence to speak in ways that meet diverse localneeds.

Or has it?My contention would be that while pedagogic ideas about teaching and learn-

ing on the one hand and sociolinguistic ideas about the sovereignty and prestigeof indigenized varieties of English on the other may have changed quite dra-matically, while the empire writes back and non-native teachers assert them-selves, assumptions about the ‘E’ in TEFL have remained curiously unaffectedby these momentous developments. In TEFL, what constitutes a valid target isstill determined with virtually exclusive reference to native-speaker norms. True,at least the perception of what constitutes ‘native speakers’ is widening, but aquestion in urgent need of exploration is just what the ‘English’ is that is beingtaught and learnt in this emerging global era, how it squares with the socio-political and socioeconomic concerns discussed in the profession, and what itsrelevance is for the subject taught in classrooms all over the world.

That this issue has not really been on the agenda so far is borne out by theway ‘English’ is talked about in the relevant literature – the default referent,implicitly or explicitly, is ENL (English as a native language):

. . . we suffer from an inferiority complex caused by glaring defects in our knowledgeof English. We are in constant distress as we realize how little we know about thelanguage we are supposed to teach. (Medgyes 1994: 40, emphases added)

I believe in the fundamental value of a common language, as an amazing worldresource which presents us with unprecedented possibilities for mutual understand-ing, and thus enables us to find fresh opportunities for international cooperation. Inmy ideal world, everyone would have fluent command of a single world language. Iam already in the fortunate position of being a fluent user of the language which ismost in contention for this role, and have cause to reflect every day on the benefitsof having it at my disposal. (Crystal 1997: viii, emphasis added)

Consider what Medgyes and Crystal3 are referring to by “English” and “thelanguage”. Of course this depends on what they regard as ‘a language’, andthere is, as Pennycook (1994: 26ff) also demonstrates, no definitive answer tothis question. But I do take it to be a general consensus that what constitutes alanguage, and in particular ‘English as a global language’, is necessarily adiscursive construct in need of deconstruction. The point I wish to make withreference to the above two extracts, then, is that the fact that these questionssimply are not problematized throws readers back on the implicit, default



referent, whether this is the one intended by the author or not. And this is oneparticular variety of English, namely that used by educated native speakerslike Crystal himself. What he is “a fluent speaker” of is English as a nativelanguage (ENL), and this variety is then bound to be understood to be “thelanguage in most contention” for the role of “a single world language”. But ofcourse like any natural language, this is full of conventions and markers of in-group membership such as characteristic pronunciations, specialized vocabu-lary and idiomatic phraseology, and references and allusions to shared experienceand cultural background. And this is precisely the reason why educated ‘non-native’ speakers of it (such as Medgyes himself ) are so resigned and defeatistabout the “glaring defects” in their knowledge of it: they cannot, by definition,be members of that native-speaker community, no matter how hard they try, nomatter how long they study.

It seems clear, then, that every time we talk and write about ‘English as aglobal language’ we quite inevitably bring into play some rather fundamentalissues. But the fact that they are fundamental has, so far at least, not meantthat they get addressed explicitly. On the contrary, the general picture is one oflack of awareness. That this constitutes a really difficult problem is probablybest illustrated with reference to writers whom nobody would suspect of native-speaker ‘tunnel vision’ and disregard for linguistic and cultural diversity.Phillipson and Skutnabb-Kangas (1999: 29) make various proposals for “chart-ing and countering Englishisation”. They suggest a number of research ques-tions focusing on the role of English in Denmark, one of which is formulatedand commented on like this:

What is the significance of senior Danish politicians, who use English with moder-ate proficiency, inevitably creating false and unintended impressions when talkingimpromptu to the ‘world’ press?

As an aside to this latter question, it should be mentioned that the four Danishexceptions to the Maastricht Treaty were hammered out at a summit in Edinburgh in1991, at the close of which the Danish Foreign Minister referred to the ‘so-calledEdinburgh agreement’, implying that no real obligation had been entered into. WhenSalman Rushdie came to Denmark in 1996 to receive an EU literature prize – anevent which was postponed because of a security scare – the Danish Prime Ministerwas asked by Rushdie whether the death threat was real or hypothetical, to which hereplied that he did not have the ‘ability’ to answer the question (a revelation thatmany Danish citizens might agree with, as the whole affair was mishandled). Areboth errors due to mother-tongue transfer? (Phillipson & Skutnabb-Kangas 1999:29f, emphases added)

This passage was rather painful to read for me, a ‘non-native’ user ofEnglish and language professional fairly conversant with some ENL varietiesbut mainly interested in English as a lingua franca (ELF), and coming as it didtowards the end of an article which contained much I agreed with, it probablyhit me with particular force. To start with, which “English” is it that Danishpoliticians use with “moderate proficiency”? As I read it, it is the same nativespeaker “English” which Crystal and Medgyes are referring to, and which is



simply assumed to be relevant in the context described. My contention would bethat it is irrelevant. To take the example of so-called, it is true of course thatENL corpora and dictionaries based on them indicate that the ‘attitudinallymarked’ use of this premodifier referred to by Phillipson and Skutnabb-Kangasis the more frequently attested one (though the other use is also attested). Butnative-speaker language use is not particularly relevant here: the Danish For-eign Minister is not a native speaker of English, and he was not speaking onbehalf of ENL speakers, nor presumably exclusively to ENL speakers. He wasusing English as a lingua franca in the way he often has occasion to use it, withinterlocutors who use it in the same way. And it is very likely indeed that suchinterlocutors would understand very well what he (presumably) meant by so-called, i.e. ‘the agreement called the Edinburgh agreement’, especially sincemany European languages have an analogous expression which can be used withthe same two meanings (German sogenannt, Italian cosiddetto, etc.). Similarly,I would claim that ability in the second example would be perfectly intelligibleto ELF interlocutors and only be perceived as odd if judged against ENL stand-ards. To call these two formulations “errors” is counterproductive at the veryleast, and evidence that ELF as a use in its own right, and ELF speakers aslanguage users in their own right, have not yet entered peoples’ consciousness,not even in the case of colleagues who have dedicated their working lives toprotecting human language rights. Phillipson and Skutnabb-Kangas are wellknown as campaigners for linguistic equity and ecology, so we can be confidentthat what they say here is not due to an attitude of superiority; rather, I suggestit results from a general problem: a conceptual gap where English as a linguafranca should have pride of place. Lest my interpretation of Phillipson andSkutnabb-Kangas’ view of ‘English’ should be perceived as overly critical, Iought to quote the endnote attached to the paragraph cited above:

The Danes are, of course, not alone in having problems with English. In the latestcommunication, in four languages, from the follow-up group preparing the revisionof the Draft Universal Declaration on Language Rights, the Catalan secretariat statesthat this key document is ‘in the way of’ being translated (=in the course of) and thatthe scientific council is ‘pretending’ to provide a forum for debate (=aiming at). It isunreasonable to expect that Danes, Catalans or other users of English as a secondlanguage use English supremely well. The dice are loaded against them, the condi-tions for communication are not symmetrical, and native speakers often seem to beunaware of this. (op.cit.: 33 fn 6; emphases added; glosses in parentheses are theauthors’)

The point I am trying to make, then, is that it is highly problematic todiscuss aspects of global English, however critically, while at the same timepassing native speaker judgements as to what is appropriate usage in ELFcontexts. The “problems” which Danes, Catalans, etc have “with English” maybe problems in the eye of the native speaker beholder, i.e. problems if you take‘English’ to be ENL. But ‘English’ does not simply transfer intact from onecontext to another – the ‘E’ in English as a Native Language is bound to be



something very different from the ‘E’ in English as a Lingua Franca, and mustbe acknowledged as such. However, this difference is still waiting to be recog-nized, explored and acted upon in much applied linguistics, and particularly inmainstream English language teaching. Widdowson (1997) offers a conceptualframework for capturing two modes of thinking about “the spread of English”which makes the fundamental differences between them quite clear:

. . . I would argue that English as an international language is not distributed, as aset of established encoded forms, unchanged into different domains of use, but it isspread as a virtual language. . . . When we talk about the spread of English, then, itis not that the conventionally coded forms and meanings are transmitted into differ-ent environments and different surroundings, and taken up and used by differentgroups of people. It is not a matter of the actual language being distributed but ofthe virtual language being spread and in the process being variously actualized. Thedistribution of the actual language implies adoption and conformity. The spread ofthe virtual language implies adaptation and nonconformity. The two processes arequite different.

And they are likely to be in conflict. Distribution denies spread. (Widdowson1997: 139f )

It seems clear, then, that in order to capture the nature of lingua francaEnglish we need to think of it as evolving out of spread, not distribution, andacknowledge the vital role and authority of ELF users as “agents of languagechange” (Brutt-Griffler 1998: 387). There has as yet, however, been no large-scale, systematic effort to record what happens linguistically in this process. Onthe contrary, the general picture we get is that there is an established Englishbeing described more and more precisely in terms of native-speaker behaviourand then distributed. This not only does not recognize necessary diversity butacts against it: “Distribution denies spread”, as Widdowson puts it in the abovequotation. This increasing precision in description is said to get closer andcloser to the reality of native-speaker language use. But it is important torealize that native-speaker language use is just one kind of reality, and one ofvery doubtful relevance for lingua franca contexts. Moreover, as long as all thedescriptive effort is geared to capturing L1 language use, the profession’s atten-tion is deflected from the increasingly urgent issues concerning the use of Eng-lish as a lingua franca, and attitudes are reinforced which are antipathetic toELF. However, I would argue that now that the right to descriptions in theirown terms has finally been recognized for nativized varieties of English, it ishigh time that we granted the same right to ELF. My contention, then, is thatwe must overcome the (explicit or implicit) assumption that ELF could possiblybe a globally distributed, franchised copy of ENL, and take on board thenotion that it is being spread, developed independently, with a great deal ofvariation but enough stability to be viable for lingua franca communication.This assumption is of course one that has to be investigated empirically, but thepoint I wish to make here is that the need to do so has not even been acknow-ledged so far, and accordingly no comprehensive effort in this direction hasbeen undertaken to date.



Conflicting tendencies

Talking about traditional TEFL aiming at ‘distributing’ English in its “establishedencoded forms” is not to say, of course, that the description of ‘E’ in TEFL hasnot moved in the last decade or two – it has moved considerably, but in the otherdirection as it were, closer and closer to the home base: linguistic descriptionsproper have been focusing on English as it is spoken and written as a firstlanguage.4 Technological developments (allied, of course, with economic interests)have made it possible to sharpen that focus, so that we can now say with precisionwhich speech acts prevail in calls to the British Telecom helpline (McEnery2000) or which features of spoken English characterize casual conversationsamong friends and acquaintances in specific parts of the UK (Carter & McCarthy1997), not to mention the precision with which written and spoken genres cannow be profiled (cf. Biber 1988). The British component of the InternationalCorpus of English is now completed; it is a corpus of a million words of spokenand written English, fully grammatically analysed, and its spoken part is “thebiggest collection of parsed spoken material anywhere” (ICE-GB website: http://www.ucl.ac.uk/english-usage/ice-gb/index.htm). Such corpora make it possi-ble to conduct extremely revealing, fine-grained analyses of, say, the rhetoricaladverb simply in present-day British English (Aarts 1996), of synchronic anddiachronic aspects of existential there (Breivik 1990) as well as studies of “vaguelanguage” (Channell 1994) and “patterns of lexis in text” (Hoey 1991).

The last quarter of the 20th century thus saw momentous developmentsand indeed a great enrichment of the study of L1 English, and the sheer scaleand sophistication of corpus-based descriptions, e.g. drawing on the BritishNational Corpus (cf. Aston & Burnard 1998), the Collins COBUILD Bank ofEnglish or the Longman-Lancaster Corpus, have revolutionized our thinkingabout what constitutes legitimate descriptions of any language. In terms ofproducts for the general public, we now have entirely empirically-based refer-ence works such as the Longman Grammar of Spoken and Written English,“Grammar for the 21st century” [flyer] (Biber et al. 1999) or the CollinsCOBUILD English Dictionary, “helping learners with real English” [cover].But the scope of descriptions of “the English language” has also widened dra-matically: while until recently the only well-documented varieties of L1 Englishwere British and North American, the International Corpus of English (ICE)encompasses over a dozen regional varieties including, for instance, Australia,East Africa, India, New Zealand and Singapore. ICE is described as “the firstlarge-scale effort to study the development of English as a world language” (ICEwebsite: http: //www.ucl.ac.uk/english-usage/ice/index.htm). But, again, it needsto be pointed out that this world language is defined in terms of speakers forwhom English is “either a majority first language . . . or an official additionallanguage” (Greenbaum 1996: 3).5 So although it is international and indeedglobal, it actually does not include a description of the use of English by themajority of its speakers, those who primarily learnt English as a lingua francafor communicating with other lingua franca speakers (cf. Graddol 1997).6



This state of affairs is reflected in the literature about teaching: there isa myriad of books and articles in the areas “English as an international lan-guage” and “intercultural communication”. Changes in the perception of therole of English in the world have significantly influenced current thinking aboutapproaches to teaching (if not necessarily the teaching itself ) and led to anincreased socio-political and intercultural awareness (e.g. Abbott & Wingard1981; Brumfit 1982; Byram & Fleming 1998; Canagarajah 1999; Gnutzmann1999; Holliday 1994; Kramsch 1993; Kramsch & Sullivan 1996; McKay forthc.;Quirk & Widdowson 1985; Smith 1981; Strevens 1980). However, as far aslinguistic models as targets for learning are concerned, these usually do notfigure as a focal concern, or matter for reflection, at all, and so, whetherexplicitly or implicitly, native-speaker models have largely remained unques-tioned. This means that the how is changing, but linked to a what that is not.Certainly no linguistically radical proposals have been put forward which wouldmatch the thrust of the important innovations which have taken place in ped-agogy. In short, no coherent and comprehensive lingua franca model hasbeen proposed so far which does justice to these changes in terms of the actuallanguage taught. This state of affairs allows the economic, social and symbolicpower of ‘native speaker English’ to be reproduced (in the sense of Bourdieu &Passeron 1970) throughout ELT institutions and practices worldwide.7

The situation that presents itself, then, is oddly contradictory and para-doxical: on the one hand, we have a very lively and prolific field of researchproducing extralinguistic treatments of how ‘English’ is – depending on thespecific researcher’s domain of interest and ideological orientation – being vari-ously spread, used, forced upon, or withheld from the world at large, coupledwith assertions of local values and the importance of intercultural communica-tion in pedagogy. On the other hand, the rapid development in computer techno-logy has opened up hitherto undreamt-of possibilities in language description. Themain research efforts in this area, however, are not expended on studying howEnglish is actually used worldwide, but instead concentrate very much on Englishas a native language. We thus have an inverse relationship between perceivedsignificance and relevance of ‘English’ in the world at large and linguistic de-scription focusing on the “ancestral home” of the language (Achebe 1975: 62).

The two contrary developments are interdependent and even reinforce oneanother: the more global the use of ‘English’ becomes, the greater the motivation,and of course the market, for descriptions of it, which, for historical and socio-economic reasons, are largely provided by the ‘Centre’. The more such products onoffer, the more these are regarded, quite rightly, as promoting the dominance of(L1) English, and thus the more forceful the attempts in (or on behalf of) the ‘Peri-phery’ to resist ‘linguistic imperialism’ (cf. e.g. Canagarajah 1999; Phillipson 1992).

English as a lingua franca: the need for description

The intellectual battles which are being fought over issues rooted in ideologicalpositions, commercial interests, ecological concerns and social identities go largely



unnoticed by the largest group of users of ‘English’: those to whom ‘English’serves on a daily basis as a lingua franca for conducting their affairs, moreoften than not entirely among so-called ‘non-native’ speakers of the language,with no native speakers present at all. These are people who have learned‘English’ as an additional language, and to whom it serves as the most usefulinstrument (for reasons variously interrogated, lamented or celebrated in ap-plied linguistics) for communication that cannot be conducted in the mothertongue, be it in business, casual conversation, science or politics – in conversa-tion, in print, on television, or on the internet. Wherever such interactions takeplace and whatever the specific motivations and uses of English as an interna-tional lingua franca, the mismatch sketched in the last two paragraphs is quitestriking: ELF speakers are usually not particularly preoccupied with the twoprevailing research foci described above, viz. ‘corpus-based description ofnative English’ and ‘linguistic imperialism’. They are not primarily concernedwith emulating the way native speakers use their mother tongue within theirown communities, nor with socio-psychological and ideological meta-level dis-cussions. Instead, the central concerns for this domain are efficiency, relevanceand economy in language learning and language use. In Kachru’s words, “thehunger for learning the language – with whatever degree of competence – issimply insatiable” (Kachru 1997: 69) This is one reason why fighting the (ab)useof ‘English’ for exerting power and domination via mainstream ELT is such anenormous task: people need and want to acquire the instrument ‘English’ what-ever the ideological baggage that comes with it – a fact acknowledged, at leastimplicitly, even in Canagarajah’s Resisting Linguistic Imperialism in EnglishTeaching (e.g. 1999: 180f ). Another reason of course is that it is not in theinterest of those who are, for want of a feasible alternative, still widely per-ceived as the ‘source’ or ‘owners’ of the commodity ‘English’ to encourage adiscussion about ethical questions and the suitability of the goods they have tooffer. And as long as ‘English’ is kept in the conceptual straightjacket of ENL,it is difficult to see how change can be pursued proactively.

However, despite the fact that discussions of ethical issues are now availablein the public domain, freely accessible and there to be taken up by anyone whochooses to, and although some scholars have been insisting for a long time that“the unprecedented functional range and social penetration globally acquiredby English demands fresh theoretical and descriptive perspectives” (Kachru1996: 906), the suitability of the descriptive and pedagogic models we are oper-ating with for the teaching and use of ELF has hardly been investigated at all.In what follows, I shall argue that it is both necessary and feasible to enquireinto a suitable model for ELF, and offer suggestions as to how this might bedone and what implications such an enquiry might have.

It would seem, then, that there is considerable scope, and hope, for large-scale, systematic research into how English is actually used as a lingua franca.However, the work actually published in this field is still extremely scarce, andvery little descriptive research has been done that could serve as a potentialbasis for formulating a curriculum for the teaching of ELF. Having said this, a



description of the phonology of English as an international language (Jenkins2000) is now available, and important work on the pragmatics of ‘non-native–non-native’ communication in English has been, and is being, conducted (e.g.Firth 1996; Meierkord 1996; House 1999; Lesznyak forthc.). James (2000)offers a rich conceptual discussion of the place of English in bi/multilingualismand makes reference to a project, currently in its pilot phase, entitled “Englishas a lingua franca in the Alpine-Adriatic region”. He also sets out hypothesesas to what findings the future analysis of this use of English by speakers ofGerman, Italian, Slovene and Friulian might yield. However, the shift in per-spective from treating, in Kasper’s words, “non-nativeness as problem” to view-ing “non-nativeness as a resource” or “non-nativeness as unattended” [becauseirrelevant] (Kasper 1997: 356f ) is a recent one in both phonology and prag-matics, and certainly not subscribed to by a majority of scholars in these fields.

It would be beyond the scope of this article to try and summarize theintriguing findings of these studies, so some examples of the kinds of insightoffered by them will have to suffice. Jenkins’ work (e.g. 1998, 2000) centresaround “a pedagogical core of phonological intelligibility for speakers of EIL”(2000: 123) which she was able to propose after establishing which pronunci-ation features impeded mutual intelligibility in her empirical studies of what sheterms ‘interlanguage talk’ among ‘non-native’ speakers of English. This proced-ure provided an empirical basis for her suggestion “to scale down the phono-logical task for the majority of learners by . . . focusing pedagogic attention onthose items which are essential in terms of intelligible pronunciation” (ibid.)and to prioritize features which constitute more relevant and more realisticlearning targets for EIL speakers. These features constitute Jenkins’ LinguaFranca Core. What I should like to emphasize in the present context is thatJenkins’ Lingua Franca Core does not include, for instance, some sounds whichare regarded, and taught, as ‘particularly English’ (and also as particularlydifficult) ones by most learners and teachers, such as the phonemes /T/ and /D/and the ‘dark l’ allophone, [¬]. That is to say that mastery of these soundsproved not to be crucial for mutual intelligibility and so various substitutions,such as /f, v/ or /s, z/ or /t, d/ for /T, D/ are permissible, and indeed also foundin some native-speaker varieties.

While phonology is a fairly ‘closed system’ (although it does have fuzzyedges), pragmatics is a more open-ended affair, and accordingly findings in thisarea as regards ELF communication are different in nature and probably shouldnot be expected to be ‘conclusive’ in the same way. As House indicates, thevolume of research in ELF pragmatics undertaken so far is minimal: “studiesof intercultural communication in the scientific community have practicallyignored ELF interactions” (1999: 74). But the findings which are beginning toemerge make it clear that there is a vast, complex and absolutely crucial areahere waiting to be explored and exploited for ELF communication. Interest-ingly, while Jenkins emphasizes the feasibility of successful communicationby means of a scaled-down phonological/phonetic repertoire, House takes amuch more sceptical stance, as reflected in the subtitle of her (1999) paper8



“Interactions in English as lingua franca and the myth of mutual intelligibility”(emphasis added). Again, I can only give a glimpse of findings here, and furtherquestions are raised by analyses carried out to date. At the most general level,an observation which has been made repeatedly is that ELF interactions oftenare consensus-oriented, cooperative and mutually supportive. For instance,a tendency has been noted to adopt a ‘Let-it-Pass’ principle, that is to say,interactants tend to gloss over utterances which cause difficulty rather thantrying to sort them out explicitly, a phenomenon Firth (1996) terms “the discurs-ive accomplishment of normality” (cf. also Meierkord 1996; Wagner & Firth1997). On the other hand, and this is what House is getting at in her subtitle,this ‘let-it-pass’ behaviour can also be interpreted as an indicator of interactants’mutual dis-attention, a “palpable lack of mutual orientation”, thus denyingeach other “the most basic social alignment between speaker and hearer” (House1999: 82). Interactants in the data analysed by her are shown to often act asinitiators only, rather than as initiators and responsive recipients, and to lack“pragmatic fluency” characterized by such features as smooth management ofturn-taking and topic-changes as well as appropriate use of pragmatic routinessuch as gambits (in the sense of Edmondson & House 1981). It has to be pointedout, however, that only a limited repertoire of interaction, notably casual con-versations and group discussions, has been analysed in this respect so far, sothat it is conceivable that further research might show the present findings to bea function of the type and purpose of the interactions investigated. Indeed, thedifferences in the analyses available to date would seem to underline the needfor a large corpus and a ‘thick description’ of the same data from variousangles. At all events, whatever ways speakers use to interact by means of alingua franca and how far they compensate for native-like conversational beha-viour is a matter for further empirical enquiry.

There is also one large-scale project focusing on the written English pro-duced by learners of ‘English’ coming from a great variety of first languagebackgrounds. This is the International Corpus of Learner English (ICLE forshort; see e.g. the contributions to Granger 1998; Granger, Hung & Petch-Tyson forthc.; Altenberg & Granger forthc.; de Haan 1998; Petch-Tyson forthc.,and more extensive studies based on this corpus, e.g. Lorenz 1999). However,the main thrust of this research enterprise is to identify characteristics of learnerEnglish from different L1 backgrounds, with the intention to facilitate compar-isons between these foreign-language productions and native-speaker writing,and so to highlight the difficulties specific L1 groups have with native English inorder to make it easier for those learners to conform to ENL if they so wish –hence the designation “Learner English”. There are also other, smaller and lessstructured learner corpora, notably those compiled by publishers big in ELT,such as the Cambridge Learners’ Corpus and the Longman Learners’ Corpus.While such projects are undoubtedly innovative and very useful in their ownterms, they are obviously quite different from my own present concern. Themain difference lies in the researchers’ orientation towards the data and thepurposes they intend the corpora to serve, namely as a sophisticated tool for



analysing learner language so as to support them in their attempts to approx-imate to native (-like) English. However, it is conceivable that some of the datain learner corpora could also contribute to a better understanding of Englishas a lingua franca. For instance, what is frequently reported as ‘overuse’ or‘underuse’ of certain expressions in learner language as compared to ENL (e.g.Chen 1998; Lorenz 1998) could also be regarded as a feature characterizingsuccessful ELF use, or the ‘deviations’ from ENL norms reported in learnercorpora research could be investigated to establish whether they can serve aspointers, or sensitizing devices, in the process of trying to profile ELF as aviable variety.

The difference in perspective between learner corpus research and my ownis an important one, and essential for the main point I am trying to make in thispaper. This difference can best be captured with reference to the notions oflearning strategies vs. communication strategies (Bialystok 1990; Corder 1981:ch. 11; Faerch & Kasper 1983).9 The students providing data for a learnercorpus have been asked, usually in an instructional setting, to produce, say, anessay in ‘English’. In the subsequent analysis, any observations about theirlanguage use are made in comparison with what native speakers would norm-ally write (as reflected in an ENL reference corpus), for instance in terms ofsimplifying, replacing, overusing, underusing or avoiding certain features. Any‘deviations’ from a native speaker norm are, then, to be seen as products oflearning strategies: a constructive way of making do with the limited linguisticresources available at a particular stage of interlanguage. The important pointto note is that the very same utterances can be regarded as communicationstrategies: evidence not of a linguistic deficit, but, if intelligible, of successfulcommunication. In principle, then, the same data could be conceptualized asentirely different kinds of evidence. However, learner corpora were not devisedwith this objective in mind, and their suitability for contributing to a descrip-tion of ELF would have to be examined with care.

Towards an ELF Corpus

So far I have been addressing what I see as an urgent (extralinguistic) need fora conceptualisation of ELF. Clearly, however, more needs to be said about the(linguistic) feasibility of such an enquiry as well as its fit with current researchparadigms. Of course, there have been various attempts in the past to do this,either as conceptually devised models of a reduced inventory as a first step,lightening the learning load as it were, from Ogden’s Basic English, which wasextremely influential in its time (e.g. Ogden 1930; see also Seidlhofer forthc.) toQuirk’s Nuclear English (1982) or as empirically derived suggestions based onmanual vocabulary counts, the most famous of these being West’s GeneralService List (1953).10 None of these, however, fulfilled the combination of cri-teria that need to be met for a viable alternative to ENL in its own right, ofwhich a very broad and substantial empirical base and a truly fresh approach



in terms of independence from the dictates of ENL would seem to be the mostimportant ones. The big opportunity which offers itself now is that it has be-come possible to take into account the considerable amount of conceptual workundertaken in the past and present while at the same time basing investigationson a large empirical foundation.

The feasibility of such a project is basically a question of methods and con-sequently has much to do with technology. But, as John Sinclair, the pioneerof corpus-based language description, so vividly demonstrates in his work, com-putational research on language has revolutionized language observation, analysisand description, in short, the whole research paradigm. And it is this, I wouldargue, which is waiting to be extended to research into ELF – needless to say, tocomplement ENL, not to replace it.

Let me suggest what it might mean to genuinely carry both the spirit and thetechnology of recent developments of language description over into the realmof ELF, to follow them through into a truly global view of English. Here isan extract from Sinclair’s introduction to his book Corpus, Concordance, Col-location (1991), in which he writes about new ways of approaching languagedescription with reference to native English. Readers are invited to engage ina thought experiment: simply imagine that what is being talked about is notnative English, but English as a Lingua Franca:

This book charts the emergence of a new view of language, and the technologyassociated with it. Over the last ten years, computers have been through severalgenerations, and the analysis of language has developed out of all recognition.

The big difference has been the availability of data. The tradition of linguistics hasbeen limited to what a single individual could experience and remember . . . Starvedof adequate data, linguistics languished – indeed it became almost totally introverted.It became fashionable to look inwards to the mind rather than outwards to society.Intuition was the key, and the similarity of language structure to various formalmodels was emphasised. The communicative role of language was hardly referred to.(Sinclair 1991: 1)

Taking the liberty to utilize this extract as an aid for reflection about, andfrom, an ELF perspective, it is fairly easy to see how “the availability of data”would make “the big difference” and allow us to focus on the “communicativerole” of ELF. When Sinclair talks about changing from looking “inwards” tolooking “outwards” he is of course referring to introspection vs. observation,but assuming this point is well taken, an advocate of ELF might be forgiven forextending this extract to an analogy: this is that “starved of adequate [ELF]data” the description of English “became almost totally introverted”, i.e. focus-ing on the use of native English only, and that it “became fashionable to lookinwards” into L1 English “rather than outwards to society”, for whom, seen ona global scale, ‘English’ means ‘English as a Lingua Franca’.

In view of all this, it seemed desirable and timely to embark on the compila-tion of a corpus of English as a Lingua Franca. For the purposes of this corpus



the term ‘lingua franca’ (cf. also Knapp & Meierkord forthc.) is understood inthe strict sense of the word, i.e. an additionally acquired language system thatserves as a means of communication between speakers of different first lan-guages, or a language by means of which the members of different speech com-munities can communicate with each other but which is not the native languageof either – a language which has no native speakers. Malmkjær (1991) explainsthe term lingua franca in terms of pidginization:

. . . if the members of two or more cultures which do not use the same language comeinto regular contact with each other over a prolonged period . . . it is probable thatthe resultant language contact will lead to the development of a pidgin language bymeans of which the members of the cultures can communicate with each other butwhich is not the native language of either speech community. A pidgin language isthus a lingua franca which has no native speakers, which is often influenced bylanguages spoken by people who travelled and colonized extensively . . . and by thelanguages of the people with whom they interacted repeatedly. (Malmkjær 1991: 81)

The point to be made, however, is that while “a pidgin language is . . . a linguafranca”, a lingua franca does not need to be reduced to a pidgin language,restricted in social role and linguistic resources, such as limited vocabulary andstylistic range, elaborated only through creolization when used as a mothertongue. Elaboration does not necessarily have to be tied to native speaker use.ELF is often used in what House (1999: 74) calls “influential networks, i.e.global business, politics, science, technology and media discourse”, and it islikely that an empirical investigation of ELF will show that a sophisticated andversatile form of language can develop which is not a native language.

The compilation of an ELF corpus is now in progress at the University ofVienna. In the current initial phase, this project is supported by Oxford Uni-versity Press (and so called the Vienna–Oxford ELF Corpus). Since the inten-tion is to capture a wide range of variation, a corpus of spoken ELF is the firsttarget, at one remove from the stabilizing and standardizing influence of writ-ing. Another important reason for concentrating on the spoken medium is thatspoken interaction is overtly reciprocal, which means that not only productionbut also reception are captured, thus allowing for observations regarding theintelligibility of what interlocutors say. For the time being, the focus is onunscripted (though partly pre-structured), largely face-to-face communicationamong fairly fluent adult speakers from a wide range of first language back-grounds whose primary and secondary education and socialization did nottake place in English. The speech events being captured include private andpublic dialogues, private and public group discussions and casual conversations,and one-to-one interviews. Ideally, speakers will be making use of ELF in alargely unselfconscious, instrumental (as opposed to identificatory) way – com-pare Hüllen’s (1992) distinction between Identifikationssprache (‘language ofidentification’) and Kommunikationssprache (‘language for communication’).At least for the first phase, it was decided to operate with a narrow definition ofELF talk. That is to say, an attempt is made to meet the following additional



criteria: no native speakers should be involved in the interaction, and theinteraction should not take place in an environment where the predominantlanguage is ‘English’, such as an ‘Inner Circle’, ENL country. The size aimedfor at the first stage is approximately half a million words (i.e. similar to thespoken part of ICE-GB), transcribed and annotated in a number of ways.11

It will thus become possible to take stock of how the speakers providing thedata actually communicate through ELF, and to begin to build a characteriza-tion of how they use, or rather co-construct, ‘English’ to do so. As a firstresearch focus, it seems desirable to complement the work already done on ELFphonology and pragmatics by concentrating on lexico-grammar and discourse,in an investigation of what (if anything), notwithstanding all the diversity, mightemerge as common features of ELF use, irrespective of speakers’ first languagesand levels of proficiency. Questions investigated will include the following: Whatseem to be the most relied-upon and successfully employed grammatical con-structions and lexical choices? Are there aspects which contribute especially tosmooth communication? What are the factors which tend to lead to ‘ripples’ onthe pragmatic surface, misunderstandings or even communication breakdown?Is the degree of approximation to a variety of L1 English always proportional tocommunicative success? Or are there commonly used constructions, lexical itemsand sound patterns which are ungrammatical in Standard L1 English but gen-erally unproblematic in ELF communication? If so, can hypotheses be set upand tested concerning simplifications of L1 English which could constitute sys-tematic features of ELF? The objective here, then, would be to establish some-thing like an index of communicative redundancy, in the sense that many of theniceties of social behaviour associated with native-speaker models and identitiesmight not be operable and certain native-speaker norms might be seen to be insuspense. Indeed, it may well be that situations occur in which ‘unilateral’approximation to native speaker norms and expectations not shared in ELFinteraction leads to communication problems, and that mutual accommoda-tion12 is found to have greater importance for communicative effectiveness than‘correctness’ or idiomaticity in ENL terms. In conducting these investigations,the large body of work already available on (native) language variation andchange, nativized varieties, pidginization and creolization as well as on simpli-fication in language pedagogy will be invaluable.

Of course, it is early days yet and all these questions will have to be formu-lated and addressed with care and circumspection. Nevertheless, I should liketo offer a brief example of the kind of enquiry I have in mind. Below is adialogue between L1 speakers of German and French respectively. They havebeen asked to choose one picture out of several options which will best serve fora campaign for a charity:

Reto (L1 German) & Stephanie (L1 French)1 R: I think on the front xx on the front page should be a picture who-which2 only makes p-people to er spend money, to the charity3 S: yes



4 R: and I think er yeah maybe5 S: I think a picture with child6 R: Yeah, child are always good to7 S: Yes8 -R: to trap people spend money9 S: Yes. I think, erm, let me see, erm . . .10 R: I don’t know . . . but maybe we should er choose a picture who gives the11 impression that this child needs needs the money or12 S: So I think, then that’s my, this one, no13 R: Yeah it’s quite happy14 S: Yeah, she’s happy er .. Maybe this one15 R: Yeah.16 S: He look very sad . . . and he has to carry heavier vase17 R: Mm, that’s right.18 S: Too heavy for him, or . . .19 R: Hm hm20 S: But also this one, even if he’s smiling21 R: Yeah, that’s right . . . And maybe this one can show that the that the22 chari-er charity can really help23 S: Uh huh24 R: and that the charity can er make a smile on a on a chil -on on a child’s25 face26 S: Yes27 R: Yeah I think this one would be28 S: A good one29 R: It would be good

. . . long pause- self-correction-R continuationxx unintelligible(Data including transcription provided by Jennifer Jenkins)13

It is obvious that the interactants are satisfied with their discussion: theyagree on their criteria and negotiate a consensus, so in that sense we can regardthis exchange as successful communication. The conversation also has a con-structive, collaborative feel to it: in contrast to the data discussed in House(1999, see above), there is ample evidence of the interactants acting as respons-ive recipients as well as initiators: the yes’s and yeah’s tend to be genuineexpressions of agreement, backchannelling is provided in the form of Hm hmand Uh huh, and there is even one instance of one speaker completing anutterance for her interlocutor (lines 27–28). But the point to be noted is thatthis communicative success comes about despite the fact that there is hardly aturn which is ‘correct’ or idiomatic by ENL standards. We find a wide range ofoddities in terms of ‘deviation’ from ENL: the unintentionally comical phrase apicture with child in line 5 (though of course only comical for someone familiar



with the ENL meaning of with child), idiosyncrasies such as makes people tospend money (line 2), to trap people spend money (line 8) and make a smile ona child’s face (lines 24–25) and what would traditionally be called ‘seriousgrammatical mistakes’, such as missing third person -s in He look very sad(line 16), wrong relative pronoun in a picture who gives the impression . . .(lines 10–11), missing indefinite article and unwarranted comparative in he hasto carry heavier vase (line 16) as well as wrong preposition (or wrong verb) into spend money to the charity (line 2).

Seen from the perspective of current mainstream ELT, this conversationcontains many ‘errors’ which most teachers would certainly consider in need ofcorrection and remediation. Despite all these, however, the exchange betweenReto and Stephanie can be regarded as an instance of successful ELF commun-ication. Of course this type of interaction relies heavily on shared context andhas a limited potential for misunderstanding and conflict, and in many situ-ations in which ELF is used such favourable conditions will not apply. Butthis caveat does not invalidate the observation that for the purpose at hand,the kind of English that is employed works, it serves the participants quiteadequately for doing the job they have to do. The investigations I have carriedout so far have confirmed that a great deal of ELF communication is conductedat roughly the level of Reto and Stephanie’s proficiency, and that quite often itis features which are regarded as ‘the most typically English’, such as 3rd per-son -s, tags, phrasal verbs and idioms, which turn out to be non-essential formutual understanding. This observation thus closely parallels Jenkins’ findingthat mastery of the sounds often perceived as ‘particularly English’, i.e. /T/ and/D/, is not crucial for ELF communication.

Of course, to most people who have experienced the use of English as alingua franca all this might seem rather obvious: we all know intuitively thatthis is how it works. But this is exactly the point I wish to make: while we(wrongly) think we have (reliable) intuitions (the fallacy so effectively exposedby recent corpus linguistics), the problem is aggravated in the case of ELFbecause, by definition, there cannot even be any native speaker intuitions aboutELF. So what we really have is impressions of ELF rather than intuitions. Itwould seem that this makes a broad empirical base on which to substantiate orindeed contradict these impressions particularly necessary.

In the teaching of English before the advent of computer-aided corpus lin-guistics, native speakers intuitively ‘knew’, but what they really said and wrotewas not captured on a large scale, and hence was not accessible for descriptionand close investigation and thus difficult to explain to learners, especially forteachers who had not grown up as speakers of the language they were teachingand thus could not even fall back on their native-speaker intuitions as theultimate yardstick. In the case of ELF, nobody has grown up as a speaker of it.One could argue that this makes the need for an empirically based descriptioneven more urgent than in the case of ENL, where at least there are nativespeakers who can serve as informants, with all their limitations. So whereas thequestion usually asked about ENL by learners and teachers is “can one say that



in English as a mother tongue?”, it would not make sense to ask the samequestion about ELF. Rather, the only really useful analogous question aboutELF would have to be an empirical one, namely “has this been said and under-stood in English as a lingua franca?” The Vienna-Oxford ELF corpus is in-tended as a first step towards addressing this question.

Conclusion

I would agree, then, with Sinclair’s observation that “the categories andmethods we use to describe English are not appropriate to the new material.We shall need to overhaul our descriptive systems” (1985: 251). I would like toadd, however, that this needs also to apply to ELF if we want to describe it:precisely the same arguments that Sinclair is making for the description ofnative-speaker language, for establishing the ‘real English of native speakers’,apply to the requirement of establishing the ‘real English of ELF speakers’.However, the vast new technological apparatus now available has not been usedfor ELF, and the reality of ELF thus not been taken into account so far.

An important difference, of course, between corpus descriptions of ENL asopposed to ELF is that in the case of the former, corpus linguistics has beenrevising and indeed revolutionizing existing descriptions, and thus impactingsignificantly on reference materials such as dictionaries and grammars. In thecase of ELF, however, the lack of a description has also meant that codificationhas been impossible to date. And here again an opportunity arises to build onthe pioneering work which has been done on indigenized varieties of English,led by Braj Kachru (e.g. 1986, 1992a). Bamgbo7e (1998) argues very forcefullyfor codification of ‘non-native varieties’ as one of “five internal factors . . .deciding on the status of an innovation” (p. 3):

I use codification in the restricted sense of putting the innovation into a written formin a grammar, a lexical or pronouncing dictionary, course books or any other typeof reference manual. . . . The importance of codification is too obvious to bebelaboured. . . . one of the major factors militating against the emergence ofendonormative standards in non-native Englishes is precisely the dearth of codifica-tion. Obviously, once a usage or innovation enters the dictionary as correct andacceptable usage, its status as a regular form is assured. (p. 4)

What I propose, then, is to consider extending Bamgbo7e’s claim to ELF andto explore the possibility of a codification of ELF with a conceivable ultimateobjective of making it a feasible, acceptable and respected alternative to ENL inappropriate contexts of use. This is, of course, a long-term project and a hugeand laborious task – an undertaking which must be carried out with extremecare, and which should not give rise to exaggerated expectations, let alonereckless premature commercial exploitation.

Once available, a description and codification of ELF use would constitute anew resource for the design of English instruction. The extent to which this



resource is used would, of course, depend on a consideration of social, culturaland educational factors which necessarily bear upon language pedagogy. Eventhough its particular realisation is at the moment impossible to predict, it iseasy to imagine the potentially huge implications this resource would carry forteacher education, curriculum design, textbooks and for how ‘English’ might betaught for lingua franca purposes where this is deemed desirable. For instance,as Jenkins so aptly puts it:

There is really no justification for doggedly persisting in referring to an item as ‘anerror’ if the vast majority of the world’s L2 English speakers produce and under-stand it. Instead, it is for L1 speakers to move their own receptive goal posts andadjust their own expectations as far as international (but not intranational) uses ofEnglish are concerned. . . . [This] also drastically simplifies the pedagogic task byremoving from the syllabus many time-consuming items which are either unteachableor irrelevant for EIL. (Jenkins 2000: 160)

But as I have said, what exactly the relevance of such a description might bewill have to be decided with reference to locally established pedagogic criteria:I would obviously not wish to claim that just because a description is availableit should determine what is taught in specific settings or for specific purposes.But it does seem likely that the conceptualisation of ELF as an alternative toENL would open up an additional repertoire of options for appropriating ‘Eng-lish’, of teaching the ‘virtual language’ (in the sense of Widdowson 1997 quotedabove) or of using ELF as a possible first step for learners in building up a basisfrom which they can then pursue their own learning in directions (ELF or ENL)which it may be impossible, and unwise, to determine from the outset. In fact,uncoupling the language from its native speakers and probing into the nature ofELF for pedagogical purposes holds the exciting, if uncomfortable, prospect ofbringing up for reappraisal just about every issue and tenet in language teach-ing which the profession has been traditionally concerned with.14

So how far any new findings will, or should, be acted upon is of course anopen question. To be realistic, a linguistic innovation which goes against thegrain of many people’s tradition and etiquette is likely to meet with a great dealof resistance due to prejudice, market forces, vested interests, cultural sensib-ilities, aesthetic arguments and practical questions. But positive perspectivesimmediately arise as well: if recent important developments in applied lin-guistics on the meta-level are matched with an empirical basis for looking atthe linguistic manifestations of ELF, this would help close the ‘conceptual gap’I have discussed and provide us with a way of ‘naming’ ELF and making clearterminological distinctions.

There are also important advantages for ENL, and ENL speakers, in this:English as used by its native speakers has hitherto been faced with the impossibleexpectation that it should be ‘all things to all people’ and the inevitable failurein this has led to it (and its speakers) being subjected to accusations includingthose of contextual inappropriacy, cultural insensitivity and political imposition.At the same time, many native speakers of English feel that ‘their language’ is



being abused and distorted through the diversity of its uses and users. If itbecomes possible to call an instance of English use ‘English as a lingua franca’,analogous to, say, ‘Nigerian English’ and ‘English English’, this acts as a powerfulsignal that they are different ‘territories’ deserving mutual respect, and withtheir own ‘legislation’. This would open up the possibility of engaging in ‘code-switching’ or at least ‘concept-switching’, and of an uninhibited acceptance ofeach use of ‘English’ in its own right – notably the appreciation of aesthetic andemotional aspects of literature, language play, rhetorical finesse, etc. Obviously,ENL would also remain intact as a target for learning in those circumstanceswhere it is deemed appropriate. Most importantly perhaps, if ELF is conceptual-ised and accepted as a distinct manifestation of ‘English’ not tied to its nativespeakers, this opens up entirely new options for the way the world’s majority ofEnglish teachers can perceive and define themselves: instead of being ‘non-native’ speakers and perennial learners of ENL, they can be competent andauthoritative users of ELF. The ‘native speaker teacher–non-native speakerteacher’ dichotomy could then finally become obsolete in ELF settings, with theprospect of abolishing a counterproductive and divisive terminology which hingeson a negative particle, and which has bedevilled the profession for too long.

Notes

I should like to thank the anonymous reviewers who commented on an earlier draft ofthis paper for the points they raised in their detailed reports, and Peter Trudgill for hisinsightful comments. I am particularly grateful to Henry Widdowson for many discus-sions, which have given me inspiration and encouragment to pursue these ideas, and thereassurance that at least one ENL speaker can live with them.

1. Cf. Gnutzmann 2000: 357: “It has been estimated that about 80 per cent of verbalexchanges in which English is used as a second or foreign language do not involvenative speakers of English (Beneke 1991)”. Crystal (1997: 54) gives the followingestimates for speakers of English in terms of Kachru’s ‘concentric circles’: InnerCircle [i.e. first language, e.g. USA, UK] 320–380 million, Outer Circle [i.e. addi-tional language, e.g. India, Singapore] 150–300 million, Expanding Circle [i.e. for-eign language, e.g. China, Russia] 100–1000 million.

2. For those who find it hard to believe that these developments are indeed quiterecent, it might be interesting to have a look at papers which document interactionsabout the ‘native–non-native’ question, such as Akoha et al. (1991), which clearlyshow how new and strange challenges to ‘native-speakerism’ were only a decadeago, even to some applied linguists.

3. I hope it is clear that I am trying to make a general point here, not suggesting anyconspirational intent on the part of Peter Medgyes and David Crystal – I picked thetwo quotations from hundreds which would have made exactly the same point. Thiswas because their books are also thematically particularly appropriate for my concernshere, and declare it as their aim to contribute to open access and equity in this area.

4. But see below for corpora of ‘learner English’.5. “Its [ICE’s] principal aim is to provide the resources for comparative studies of the

English used in countries where it is either a majority first language . . . or anofficial additional language. In both language situations, English serves as a means



of communication between those who live in these countries” (Greenbaum 1996: 3).“Excluded from ICE is the English used in countries where it is not a medium forcommunication between natives of the country” (p. 4).

6. See also Kachru’s criticism of the Cambridge International Dictionary of Englishfor using the term ‘international’ for referring to “America, Britain and Australia”(1997: 70f.). The term “international English” is usually not taken to include ‘Expand-ing Circle’ English (cf. e.g. Todd & Hancock 1986; Trudgill & Hannah 1995).

7. In an attempt to counteract this mechanism of (unwitting) reproduction, I will useinverted commas for any mention of ‘English’ when co- and context do not indicatewhich variety is being talked about (such as ENL, ELF or an indigenized ‘OuterCircle’ variety), as a reminder that any general reference to the denomination ‘Eng-lish’ has to be regarded as provisional.

8. I focus on House (1999) here because this paper summarizes and discusses the stateof the art in the pragmatics of ELF.

9. It has to be pointed out that the terminology in the literature about ‘strategies’ is farfrom unified, cf. the overview in Bialystok (1990).

10. Crystal (e.g. 1997: 136ff.) foresees the emergence of “a new form of English – let usthink of it as ‘World Standard Spoken English’ (WSSE).” He adds, though, that“WSSE is still in its infancy. Indeed, it has hardly been born” and that it seemslikely that the variety which will be “most influential in the development of WSSE”is US English.

11. The texts are being transcribed orthographically and marked for speaker turns,pauses and overlaps, and provided with contextual notes and notes about para-linguistic features such as laughter. Part-of-speech tagging and syntactic parsing areto be added. A fairly basic system for marking for prosody is being worked out andpiloted. It is currently not planned to provide a phonetic transcription of thesetexts, but it is hoped that sound files can be made available in the longer term.

12. Accommodation (in the sense of Giles & Coupland 1991) was found to be an importantfactor in Jenkins’ (2000) study, and lack of it may have contributed to the impres-sion of “mutual dis-attention” House (1999: 82) got in the analysis of her data.

13. All the features of ELF interaction highlighted here have also been found in my owndata, but I do not have any interactions as yet which entirely consist of such shortturns as the present example, and therefore decided on the above extract as themost compact illustration available.

14. Kramsch (1999: 142) sounds a timely note of caution against a premature and naïveeuphoria about ‘global access’: “Global access to English, like global access to theInternet, facilitates communication, it does not necessarily facilitate understanding.In fact, it requires an additional effort in discursive reflexivity and linguistic cir-cumspection to overcome the illusion of sameness created by the use of a commonlanguage. A pedagogy of English as a global language can capitalize on the outsidenessof the local, non-native speaker to foster an understanding based not only on orate usesof language but on a critical reflection that can only be acquired through literacy.”

References

Aarts, B. (1996) The rhetorical adverb simply in present-day British English.In C. Percy, C. Meyer & I. Lancashire, Synchronic corpus linguistics.Amsterdam: Rodopi. 59–68.

Abbott, G. & P. Wingard (eds.) (1981) The teaching of English as an interna-tional language. Glasgow & London: Collins.

Achebe, C. (1975) Morning yet on creation day. New York: Doubleday.



Akoha, J., Z. Ardo. J. Simpson, B. Seidlhofer & H.G. Widdowson (1991)Nationalism is an infantile disease. (Einstein) What about native-speakerism?BAAL Newsletter 39: 21–26.

Altenberg, B. & S. Granger (2001) The grammatical and lexical patterning ofmake in native and non-native student writing. Applied Linguistics 22.2:173–94.

Ammon, U., K. Mattheier & P. Nelde (eds.) (1994) English only? In Europa /in Europe / en Europe. Sociolinguistica 8. Tübingen: Niemeyer.

Aston, G. & L. Burnard (1998) The BNC handbook. Edinburgh University Press.Bailey, R. & M. Görlach (eds.) (1982) English as a world language. Ann Arbour,

MI: University of Michigan Press.Bamgbo7e, A. (1998) Torn between the norms: innovations in world Englishes.

World Englishes 17.1: 1–14.—, A. Banjo & A. Thomas (eds.) (1995) New Englishes. A West African per-

spective. Ibadan: Mosuro & The British Council.Beneke, J. (1991) Englisch als lingua franca oder als Medium interkultureller

Kommunikation. In R. Grebing, Grenzenloses Sprachenlernen. Berlin:Cornelsen. 54–66.

Bhatia, V. (1997) Introduction: genre analysis and world Englishes. WorldEnglishes 16.3: 313–9.

Bialystok, E. (1990) Communication strategies. A psychological analysis ofsecond language use. Oxford: Blackwell.

Biber, D. (1988) Variation across speech and writing. Cambridge UniversityPress.

—, S. Johansson, G. Leech, S. Conrad & E. Finegan (1999) Longman Gram-mar of spoken and written English. Harlow, Essex: Pearson.

Bourdieu, P. & J.-C. Passeron (1970) Reproduction in education, culture andsociety. Transl. R. Nice, London: Sage.

Braine, G. (ed.) (1999) Non-native educators in English language teaching.Mahwah, NJ: Erlbaum.

Breivik, L.-E. (1990) Existential ‘there’: a synchronic and diachronic study(2nd edition). Oslo: Novus.

Brumfit, C. (ed.) (1982) English for international communication. Oxford:Pergamon.

Brutt-Griffler, J. (1998) Conceptual questions in English as a world language.World Englishes 17.3: 381–92.

Brutt-Griffler, J. & K. Samimy (1999) Revisiting the colonial in the postcolonial:critical praxis for nonnative-English-speaking teachers in a TESOL pro-gram. TESOL Quarterly 33.3: 413–31.

Byram, M. & M. Fleming (eds.) (1998) Language learning in interculturalperspective. Cambridge University Press.

Canagarajah, S. (1999) Resisting linguistic imperialism in English teaching.Oxford University Press.

Carter, R. & M. McCarthy (1997) Exploring spoken English. Cambridge Uni-versity Press.



Chen, H. (1998) Underuse, overuse, and misuse in Taiwanese EFL learnercorpus. In S. Granger & J. Hung, Proceedings of the First InternationalSymposium on Computer Learner Corpora, Second Language Acquisitionand Foreign Language Teaching. The Chinese University of Hong Kong.

Channell, J. (1994) Vague language. Oxford University Press.Cook, V. (1999) Going beyond the native speaker in language teaching. TESOL

Quarterly 33.2: 185–209.Corder, S.P. (1981) Error analysis and interlanguage. Oxford University

Press.Crystal, D. (1997) English as a global language. Cambridge University Press.de Haan, P. (1998) How native-like are advanced learners of English? In

A. Renouf, Explorations in corpus linguistics. Amsterdam: Rodopi. 55–65.Edmondson, W. & J. House (1981) Let’s talk and talk about it. A pedagogical

interactional grammar of English. München: Urban & Schwarzenberg.Faerch, C. & G. Kasper (1983) Strategies in interlanguage communication.

London: Longman.Firth, A. (1996) The discursive accomplishment of normality. On ‘lingua franca’

English and conversation analysis. Journal of Pragmatics 26.3: 237–59.Giles, H. & N. Coupland (1991) Language: contexts and consequences. Milton

Keynes: Open University Press.Gnutzmann, C. (ed.) (1999) Teaching and learning English as a global lan-

guage. Tübingen: Stauffenburg.— (2000) Lingua franca. In M. Byram, The Routledge encyclopedia of language

teaching and learning. London: Routledge. 356–9.Graddol, D. (1997) The future of English? London: British Council.Granger, S. (ed.) (1998) Learner English on computer. London: Longman.—, J. Hung & S. Petch-Tyson (eds.) (forthcoming) Computer learner corpora,

second language acquisition and foreign language teaching.Greenbaum, S. (ed.) (1996) Comparing English worldwide. The international

corpus of English. Oxford: Clarendon.Holliday, A. (1994) Appropriate methodology and social context. Cambridge

University Press.House, J. (1999) Misunderstanding in intercultural communication: inter-

actions in English as a lingua franca and the myth of mutual intelligibility.In C. Gnutzmann, Teaching and learning English as a global language.Tübingen: Stauffenburg. 73–89.

Hüllen, W. (1992) Identifikationssprachen und Kommunikationssprachen.Zeitschrift für Germanistische Linguistik 20.3: 298–317.

James, A. (2000) English as a European lingua franca. Current realities andexisting dichotomies. In J. Cenoz & U. Jessner, English in Europe. Theacquisition of a third language. Clevedon: Multilingual Matters. 22–38.

Jenkins, J. (1998) Which pronunciation norms and models for English as anInternational Language? ELT Journal 52.2: 119–26.

— (2000) The phonology of English as an international language. Oxford Uni-versity Press.



Kachru, B. (1986) The alchemy of English: the spread, functions and models ofnon-native Englishes. Oxford: Pergamon. [Reprinted 1991, University ofIllinois Press, Urbana, Il.]

— (ed.) (1992a) The other tongue (2nd edition). Urbana/Chicago: University ofIllinois Press.

— (1992b) Models for non-native Englishes. In B. Kachru, The other tongue(2nd edition). Urbana/Chicago: University of Illinois Press.

— (1996) English as lingua franca. In H. Goebl, P. Nelde, Z. Stary & W.Wölck, Kontaktlinguistik. Contact linguistics. Linguistique de contact.Vol. 1. Berlin: De Gruyter. 906–13.

— (1997) World Englishes and English-using communities. Annual Review ofApplied Linguistics 17: 66–87. Cambridge University Press.

Kasper, G. (1997) Beyond reference. In G. Kasper & E. Kellerman, Com-munication strategies: psycholinguistic and sociolinguistic perspectives.London: Longman. 345–60.

Knapp, K. & C. Meierkord (eds.) (forthcoming) Lingua franca communica-tion. Frankfurt/Main: Lang.

Kramsch, C. (1993) Context and culture in language teaching. Oxford Univer-sity Press.

— (1998) The privilege of the intercultural speaker. In M. Byram & M. Fleming,Language learning in intercultural perspective. Cambridge University Press.16–31.

— (1999) Global and local identities in the contact zone. In C. Gnutzmann,Teaching and learning English as a global language. Stauffenburg. 131–43.

— & P. Sullivan (1996) Appropriate pedagogy. ELT Journal 50.3: 199–212.Lesznyak, A. (forthcoming) Untersuchungen zu Besonderheiten der

Kommunikation im Englischen als Lingua Franca. Ph.D. Dissertation,Universität Hamburg.

Lorenz, G. (1998) Overstatement in advanced learners’ writing: stylistic aspectsof adjective intensification. In S. Granger, Learner English on computer.London: Longman. 53–66.

— (1999) Adjective intensification – learners versus native speakers. A corpusstudy of argumentative writing. Amsterdam/Atlanta: Rodopi.

Malmkjær, K. (1991) The linguistics encyclopedia. London/New York: Routledge.McArthur, T. (1998) The English languages. Cambridge University Press.McEnery, T. (2000) Speakers, hearers and annotation – why it’s difficult to do

pragmatics with corpora. Paper given on 9 February, Institute of EnglishStudies, University of London.

McKay, S. (forthcoming) Teaching English as an international language: re-thinking goals and approaches. Oxford University Press.

Medgyes, P. (1994) The non-native teacher. London: Macmillan.Meierkord, C. (1996) Englisch als Medium der interkulturellen Kommunikation.

Untersuchungen zum non-native-/non-native speaker – Diskurs. Frankfurt/Main: Lang.



Ogden, C. (1930) Basic English. A general introduction with rules and gram-mar. London: Kegan Paul.

Pennycook, A. (1994) The cultural politics of English as an international lan-guage. London: Longman.

— (1999) Introduction: critical approaches to TESOL. TESOL Quarterly 33.3:329–48.

Petch-Tyson, S. (forthcoming) Demonstrative expressions in argumentative dis-course – a computer-based comparison of non-native and native English. InS. Botley & T. McEnery, Corpus-based and computational approaches todiscourse anaphora.

Percy, C., C. Meyer & I. Lancashire (eds.) (1996) Synchronic corpus linguistics.Amsterdam: Rodopi.

Phillipson, R. (1992) Linguistic imperialism. Oxford University Press.— & T. Skutnabb-Kangas (1999) Englishisation: one dimension of globalisation.

AILA Review 13: 19–36.Rampton, B. (1990) Displacing the ‘native speaker’: expertise, affiliation and

inheritance. ELT Journal 44.2: 97–101.Quirk R. (1982) International communication and the concept of Nuclear Eng-

lish. In C. Brumfit, English for international communication. Oxford:Pergamon. 15–28.

— & H.G. Widdowson (1985) (eds.) English in the world: teaching and learn-ing the languages and literatures. Cambridge University Press.

Seidlhofer, B. (1999) Double standards: teacher education in the ExpandingCircle. World Englishes 18.2: 233–45.

— (forthcoming) The shape of things to come? Some basic questions. InK. Knapp & C. Meierkord, Lingua franca communication. Frankfurt/Main: Lang.

Sinclair, J. (1985) Selected issues. In R. Quirk & H.G. Widdowson, English inthe world: teaching and learning the languages and literatures. CambridgeUniversity Press. 248–54.

— (1991) Corpus, concordance, collocation. Oxford University Press.Smith, L. (ed.) (1981) English for cross-cultural communication. London:

Macmillan.— & M. Forman (eds.) (1997) World Englishes 2000. Honolulu, Hawai’i: Col-

lege of Languages, Linguistics and Literature and the East-West Center.Todd, L. & I. Hancock (1986) International English usage. London: Croom

Helm.Trudgill, P. & J. Hannah (1995) International English: a guide to varieties of

standard English (3rd edition). London: Arnold.Wagner, J. & A. Firth (1997) Communication strategies at work. In G. Kasper

& E. Kellerman, Communication strategies: psycholinguistic and sociolin-guistic perspectives. London: Longman. 323–44.

West, M. (1953) A general service list of English words: with semantic frequen-cies and a supplementary word-list for the writing of popular science andtechnology. London: Longmans.



Widdowson, H.G. (1994) The ownership of English. TESOL Quarterly 28.2:377–89.

— (1997) EIL, ESL, EFL: global issues and local interests. World Englishes16.1: 135–46.

[Received 10/5/01]

Barbara SeidlhoferInstitut für Anglistik und Amerikanistikder Universität WienUniversitätscampus AAKHSpitalgasse 2–4A-1090 Vienna, Austriae-mail: [email protected]

Closing a conceptual gap: the case for a description of ...

Documents

Transcript of Closing a conceptual gap: the case for a description of ...